Transcription factor Cdx4 and transcriptional coregulator menin are essential for Hoxa9 expression and normal hematopoiesis. However, the precise mechanism underlying Hoxa9 regulation is not clear.
Methods and Findings
Here, we show that the expression level of Hoxa9 is correlated with the location of increased trimethylated histone 3 lysine 4 (H3K4M3). The active and repressive histone modifications co-exist along the Hoxa9 regulatory region. We further demonstrate that both Cdx4 and menin bind to the same regulatory region at the Hoxa9 locus in vivo, and co-activate the reporter gene driven by the Hoxa9 cis-elements that contain Cdx4 binding sites. Ablation of menin abrogates Cdx4 access to the chromatin target and significantly reduces both active and repressive histone H3 modifications in the Hoxa9 locus.
Citation: Yan J, Chen Y-X, Desmond A, Silva A, Yang Y, Wang H, et al. (2006) Cdx4 and Menin Co-Regulate Hoxa9 Expression in Hematopoietic Cells. PLoS ONE 1(1): e47. https://doi.org/10.1371/journal.pone.0000047
Academic Editor: Christopher Arendt, Sanofi-Aventis, United States of America
Received: September 26, 2006; Accepted: October 10, 2006; Published: December 20, 2006
Copyright: © 2006 Yan et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was in part supported by NIH grant (CA100912 to XH) and the Burroughs Wellcome Career Award (XH).
Competing interests: The authors have declared that no competing interests exist.
Homeo-box-containing transcription factors (Hox proteins) play a pivotal role in normal differentiation and expansion of hematopoietic cells , . Overexpression of several Hox genes such as Hoxa9 induce leukemia reminiscent of mixed lineage leukemia (MLL) . The MLL gene is frequently targeted by chromosomal translocations in acute lymphoid and myeloid leukemias (ALL and AML, respectively) , resulting in generation of various MLL-fusion proteins with the amino portion of MLL fused in frame with one of over 50 different potential fusion partners. Many MLL fusion proteins cause dysregulation of Hox genes including Hoxa9. Both human and murine leukemias triggered by MLL-AF9 are most frequently myeloid in phenotype –.
Menin was originally identified as a tumor suppressor encoded by Men1 that is mutated in the human inherited tumor syndrome, multiple endocrine neoplasia type 1 (MEN1) , . It interacts with a variety of transcriptional factors and MLL proteins –. Recently, we and others have shown that menin regulates hematopoiesis and myeloid transformation by influencing Hox gene expression , . Menin and MLL fusion proteins are required for persistent expression of 5′ Hoxa cluster genes, including Hoxa5 to Hoxa10 . Despite the advances in understanding the oncogenic contributions of the menin-MLL complex, precisely how expression of Hoxa cluster genes is regulated in hematopoietic cells remains unclear.
The vertebrate Cdx genes (Cdx1, Cdx2 and Cdx4 in the mouse) encode homeodomain transcription factors related to the Drosophila caudal genes. They belong to the ParaHox gene family and share common ancestry with the Hox genes . In zebrafish, a Cdx4 mutant, kggtv205, causes an early defect in hematopoiesis and aberrant expression of multiple Hox genes, particularly Hoxb6b and Hoxa9a . Ectopic expression of Cdx4 induced a pronounced expansion of multi-lineage progenitors from differentiating mouse embryonic stem cells, including an increase in granulocyte-erythroid-macrophage-megakaryocyte colonies and in granulocyte-macrophage colonies, as compared with control cells . These data indicate that Cdx4 is essential for Hoxa9 regulation in normal hematopoiesis.
There is substantial data demonstrating an essential role for Hoxa9 protein in both normal hematopoiesis  and leukemic transformation induced by certain MLL fusion proteins . Hoxa9 overexpression is a common feature of acute myeloid leukemia –. Although both menin and Cdx4 have been shown to participate in regulating Hoxa9 gene transcription and hematopoiesis, little is known as to whether or how Cdx4 interacts with the Hoxa9 locus to regulate its expression, and whether menin and Cdx4 crosstalk to regulate expression of Hoxa cluster genes.
In this study, we used two MLL-AF9 transformed cell lines, AT1 and AR1, as a model system to investigate Hoxa9 regulation. We showed that histone modifications, especially methylation of histone H3, play an important role in determining the level of Hoxa9 expression, and menin is crucial for both active and repressive histone modifications in the Hoxa9 locus. We also demonstrated a crosstalk between menin and Cdx4 in co-regulating expression of Hoxa9 gene. These results uncovered a functional link among Cdx4, menin and histone modifications of histone H3 in maintaining 5′ Hoxa cluster transcription, thus providing novel insights into our understanding of regulation of Hox gene expression.
Distinct levels of Hox gene expression in two MLL-AF9 transformed cell lines
AT1 and AR1 cell lines were both transformed by transduction of MLL-AF9 fusion genes, however, compared to AT1, AR1 cells were cultured easily and grew faster (S Fig. 1A). The expression levels of Hoxa5, Hoxa7 and Hoxa9 in AR1 cells, as measured by qPCR, were 20 to 40 fold higher than those in AT1 cells. The Hoxa10 level in AR1 cells was even 105 fold higher than in AT1 cells (S Fig. 1B), while the expression of Hoxa1 and Hoxa13 was not detectable in AT1 cells. These results suggest that expression of multiple Hox genes in AR1 cells is generally higher than that in AT1 cells, consistent with a more aggressive phenotype of AR1 cells (S Fig. 1C).
(A) Diagram of 5′ Hoxa cluster genes, from Hoxa5 to Hoxa11, showing the location of the amplicons used for quantitative Chromatin Immunoprecipiation assay (qChIP).
The enrichment of H3K4 methylation was determined by qChIP in AT1 cells (B), AR1 cells (C) and MEF26 cells (D).
(E) Hoxa9 expression levels in the indicated cell lines as determined by qPCR.
The relative expression level of Hoxa9 gene was calculated relative to that in MEF26 cells.
More importantly, the morphology and expression profile of 5′ Hoxa cluster genes in AT1 and AR1 cells are strikingly similar to the previously published two distinct stages of MLL-AF9 transformed hematopoietic cells: preleukemic and leukemic stages , , . In those studies, the authors found that MLL-AF9 fusion induced generation of preleukemic cells followed by overt leukemic cells . In contrast to leukemic cells, preleukemic cells possessed a high differentiation potential and expressed a lower level of 5′ Hoxa cluster genes , . Mouse embryonic fibroblast (MEF) cells were often used to study histone modifications in Hoxa9 gene expression , . The expression level of Hoxa9 in AT1 cells was only two fold higher than in MEF cells (Fig. 1E). Apparently AT1 cells have not transformed into overt leukemic cells but possessed differentiation potential associated with an early stage of MLL-AF9 transformation, similar to the preleukemic cells . Therefore, utilization of AR1 and AT1 cell lines would facilitate our understanding of the altered regulation of Hox genes in various stages of leukemia.
Multiple functional elements spread over the regulatory sequence of Hoxa9 gene
Initiation of Hox gene transcription requires interaction among numerous factors recruited to the target gene, particularly to the multiple functional elements, so as to establish their precise domain boundaries in the developing tissue . The Hoxa9 regulatory region may serve as binding sites for various transcription factors and co-regulators to regulate the magnitude of Hoxa9 expression. To identify the regulatory elements in the Hoxa9 gene, we employed a luciferase reporter assay to quantitatively analyze the cis-elements in the Hoxa9 locus to regulate its expression. Various cis-elements ranging from −1222 to +1226 of the Hoxa9 locus were inserted upstream of the luciferase reporter to generate reporters pA through pF (S Fig. 2A). The effects of each regulatory element on reporter gene activity were tested in MEFs (S Fig. 2B). Luciferase assays identified a strong activating element located at 5′ promoter region, B, and two other activating elements within the 5′ untranslated region (5′-UTR), E and F. A repressive fragment, as represented by elements C and D, is located between the two activating elements, B and E (S Fig. 2B). These results suggest that both positive and negative regulators are involved in regulating Hoxa9 transcription through distinct cis-elements in the Hoxa9 locus.
Active histone markers including H4K16Ace, H3K4M2 and H3K4M3, were indicated as grey bars.
Repressive markers including H3K4M1, H3K9M2 and H3K27M2 were indicated as black bars.
(A) AT1 cells maintain both active and repressive histone modifications at amplicon 2.
(B) AR1 cells show aberrant high expression of active histone modifications but low level of repressive histone modifications at amplicon 2.
(C) and (D) show that MEF26 cells contain both active and repressive histone markers at amplicons 2 and 4, respectively.
The level of Hoxa9 expression correlates with the location of the increased lysine 4-trimethylated histone H3 in the Hoxa9 locus
We and others have separately demonstrated that the menin and MLL complex ,  and the SUZ12-EZH2 (polycomb proteins)  interact with Hoxa9 regulatory regions. MLL confers histone H3 lysine 4 (H3K4) methlyation, while SUZ12-EZH2 mediates H3K27 methylation , . However, precisely how histone H3 modifications are distributed among positive and negative elements in the Hoxa9 locus is poorly understood. Having identified the potentially active and repressive element, we used quantitative chromatin immunoprecipitation (qChIP) to map a variety of histone modifications in the Hoxa9 locus, as represented by amplicons 1–4 (Fig. 1A) in three different cell lines, each expressing a distinct level of Hoxa9 (Fig. 1E).
The patterns of H3K4 methylation across the regulatory region of Hoxa9 are distinct among the three cell lines (Fig. 1B to D). In AT1 cells, which express a medium level of Hoxa9 among the three cell lines, there were marked enrichments of multiple histone modification markers including H3K4 trimethylation (H3K4M3), H3K4 dimethylation (H3K4M2) and H3K4 monomethylation (H3K4M1) that peaked at amplicon 2 and extended across amplicons 1, 3 and 4. The enriched level of H3K4M3, H3K4M2 and H3K4M1 accounted for 4–9%, 2–4% and 6–13% of the input among the four regions, respectively (Fig. 1B). In contrast, a different pattern of H3K4 modifications was detected in AR1 cells, which express a much higher level of Hoxa9, 41 fold higher than in AT1 cells (Fig. 1E). Although the level of H3K4M3 at amplicons 3 and 4 was comparable to that in AT1 cells, it gradually increased toward 5′ amplicons 1–2, culminating at amplicon 1 to 28% of the input in AR1 cells, as compared to 6.5% in AT1 cells (Fig. 1C). On the other hand, H3K4M1 and H3K4M2 in AR1 cells were either quite low or barely detectable. To compare our results of histone modifications with previously published data with MEFs , , we also measured H3K4 modifications among the 4 amplicions in MEFs which express a low level of Hoxa9 (Fig. 1E). In contrast to AR1 cells, no H3K4 methylations were detected at amplicon 1, while H3K4M3 was gradually increased from 6% to 14% from amplicons 2 to 4. Enrichment of H3K4M1 and H3K4M2 mimicked the trend of H3K4M3, albeit at a relatively lower level. Although H3K4 methylations in MEF26 cells were maintained at a relatively high level at amplicon 4, the expression of Hoxa9 gene was low (Fig. 1E). These results demonstrate that H3K4 trimethylation at the 5′ sequence examined in the Hoxa9 locus is closely correlated with a high level of Hoxa9 expression, suggesting the importance of the location of methylated H3K4 in upregulating Hoxa9 expression.
Active and repressive histone modifications coexist in the Hoxa9 regulatory region
It has been reported that H3K4M2 and H3K4M3, and acetylation of histone 4 lysine 16 (H4K16Ace) are correlated with transcriptionally active genes, whereas H3K4M1, dimethylation of histone 3 lysine 9 (H3K9M2) and dimethylation of histone 3 lysine 27 (H3K27M2) are associated with transcriptional silencing –. Although methylation of histone H3K4, H3K9 and H3K27 are important for maintenance of normal Hoxa9 expression , , ,  whether these positive and repressive histone modifications co-exist in Hoxa9 regulation is unclear.
As noticed in AT1 cells, both active (H3K4M2, H3K4M3 and H4K16Ace) and repressive (H3K4M1, H3K9M2 and H3K27M2) histone modification markers were differentially enriched at the 5′ promoter region (amplicon 2) (Fig. 2A). In contrast, in AR1 cells H3K4M3 was highly increased but other modifications were suppressed (Fig. 2B). Similar to AT1 cells, MEF26 cells possessed both active and repressive histone modifications at both amplicons 2 and 4 (Fig. 2C, D). Enrichment of both active and repressive chromatin at the same region suggests that they may antagonize or cooperate to establish a dynamic balance. This balance could be useful for maintaining the level of Hoxa9 expression. An alteration in this balance could result in deregulation of Hoxa9 expression. This is consistent with hyperexpression of Hoxa9 in AR1 cells in which the repressive histone modifications in the Hoxa9 locus markedly reduced (Fig. 1C, 1E and 2B). It is likely that the co- enrichment of repressive histones moderates the high expression of Hoxa9.
Menin is required for active and repressive histone H3 modifications in the Hoxa9 locus and the expression of 5′ Hoxa cluster genes
To evaluate the role of menin in histone modifications at various locations of the Hoxa9 locus, we treated AT1 cells with either control DMSO or 4-OH tamoxifen (4OHT) to excise floxed Men1, followed by qChIP assay. 4OHT markedly reduced menin expression (Fig. 3D). In menin-expressing AT1 cells, H3K4M3 associated with each of the 4 amplicons, reaching 6.5–9% of the input at amplicons 1–2 (Fig. 3A, right panel). Other histone modifications, H3K4M2 and H4K16Ace, were also detectable at amplicons 1–4, albeit at a lower level (Fig. 3A, right panel). In contrast, Men1 excision (Men1Δ/Δ) markedly reduced all three active histone modifications (Fig. 3A, right panel). This result is in agreement with the previous observation that menin facilitates trimethylation of H3K4, activates Hoxa9 expression , and also links menin's function to H4K16Ace. Notably, Men1 excision also markedly reduced all three repressive histone modifications at the 4 amplicons of the Hoxa9 locus (Fig. 3B).
Menin excision (Men1Δ/Δ) dramatically reduced active (A) and repressive (B) histone modifications.
The highly enriched peak of H3K4M3 and H3K4M2 were almost completely abrogated.
(C) Menin excision also reduced the expressions of Hoxa5, Hoxa7, Hoxa9 and Hoxa10.
No expression of Hoxa1 and Hoxa13 was detectable.
(D) 4-OH Tamoxifen (4-OHT) treatment substantially reduced menin expression as compared to non-tamoxifen treated AT1 cells.
Mouse β-actin was used as control.
To verify the functional links between menin-associated histone 3 modifications and Hox gene expression, we conditionally excised menin in AT1 cells and examined menin's influence on the expression of 5′ Hoxa cluster. In keeping with reported results that Hoxa5 –a10 are co-expressed in MLL-AF9 transformed cells , , , menin excision significantly reduced the expression of Hoxa9 (7 fold), Hoxa7 (5 fold), Hoxa5 (2 fold) and Hoxa10 (3 fold) (Fig. 3C). These results suggest that menin is required to maintain the expression of 5′ Hoxa cluster genes in hematopoietic cells through mediating multiple methylations at histone 3.
Cdx4 is required to regulate the expression of 5′ Hoxa cluster genes at a certain stage of hematopoiesis
Previous studies have shown that menin and Cdx4 induce hematopoiesis at least in part through regulating Hox genes, especially Hoxa9 , . In agreement with the finding that Cdx4 is required for Hoxa9 and Hoxa7 expression , our retrovirus-based Cdx4 shRNAs experiments showed that Cdx4 knockdown decreased the expression of 5′Hoxa cluster genes, Hoxa5- a10; particularly Hoxa9 expression was reduced by 38 fold relative to the control (data not shown). These data suggested that the Cdx4 gene is crucial for 5′Hoxa cluster expression in AT1 cells, and especially for Hoxa9 regulation. As Cdx4 is almost absent in AR1 cells (Fig. 4A), we infected AT1 and AR1 cells by Cdx4-expressing retrovirus. The same vector containing IRES-GFP but not Cdx4 cDNA was used as control. We noticed that ectopic expression of Cdx4 increased the expression of Hoxa cluster genes 2 fold in AT1 cells (Fig. 4B) but did not change the expression of these genes in AR1 cells (Fig. 4C). Menin expression was not influenced by expression level of Cdx4 in both AT1 cells and AR1 cells (Fig. 4 B, C). Given that AT1 cells express a lower level of Hoxa genes and have the potential to differentiate into myelomonocytes and myeloneutrophils (S Fig. 1C), we concluded that Cdx4 gene may be temporarily required to establish and modulate Hox gene expression at specific stages of hematopoiesis, possibly during myeloid transformation , . However, Cdx4 is no longer required once Hoxa cluster genes have been highly activated in cells such as AR1 cells. Consistent with this notion, in AR1 cells Hoxa9 expression was 41 fold higher than that in AT1 cells but Cdx4 expression was almost absent (Fig. 4A).
Cdx4 and Menin co-regulate expression of the reporter activity driven by Hoxa9 cis-elements
We further determined whether Cdx4 cooperates with menin in Hox expression. To this end, we used Hoxa9 gene as a model to evaluate the effect of menin and Cdx4 on reporter gene activity (Fig. 5A). We cotransfected four reporters (pB, pD, pE and pF) containing different Hoxa9 regulatory elements with menin and/or Cdx4-expressing constructs into 293 cells. As shown in Fig.5B, when reporter pB was cotransfected with Cdx4 and/or menin, its activity was induced by Cdx4 and enhanced by co-expressed menin (Fig. 5B). Neither menin nor Cdx4 alone had significant effects on reporters pD and pF (Fig. 5B). Interestingly, cotransfecting both menin and Cdx4 with reporter pE yielded an additive activation of the reporter expression by 3.5 fold higher than either menin (1.3 fold) or Cdx4 (2.5 fold) alone (Fig. 5B). Thus, the luciferase assays suggest that Cdx4 could directly activate the activity of the Hoxa9 elements, and menin may enhance its activation.
(A) Diagram of Hoxa9 locus, showing the locations of amplicons and the corresponding reporter elements B, D, E and F.
(B) Co-expression of menin and Cdx4 with reporters in 293 cells induced the activity of reporters pB and pE.
The reporter activity was arbitrarily calculated relative to vector control.
(C) Western blot shows that in vitro synthesized and transiently expressed Cdx4 was specifically detected by anti-Cdx4 antibodies.
Cdx4 protein specifically binds to two sites in the Hoxa9 regulatory region in vitro
To determine whether Cdx4 directly interacts with Hoxa9 regulatory elements, we performed sequence homolog search for the consensus binding sites (TTTATA/G) for Cdx4 . The search revealed three Cdx4 binding sites (CBS): one in the 5′ part of element B and the other two closely located within element E (Fig. 6A and B). We then performed gelshift assays to establish whether Cdx4 can bind these sites. Of six wild type oligonucleotide sequences used as probes (see material and methods), only two probes (probe 2 and 6) were specifically bound by in vitro synthesized Cdx4 protein (A.S. data not shown). We then focused on these two probes. Both probes 2 and 6 bound the in vitro translated Cdx4 (Fig. 6C, lanes 3 and 17), but not the control lysate (lanes 2 and 16). Moreover, Cdx4/probe 2 complex was abolished by incubation with an anti-Cdx4 antibody but not by control IgG (lanes 4 and 6), in agreement with the previous report , suggesting that Cdx4/probe 2 interaction or stability of the complex was diminished by the antibody. However, Cdx4/probe 6 complex was further supershifted by the same anti-Cdx4 antibody, but not by the control IgG (Fig. 6C, lanes 18 and 20). Since probe 6 contains two CBS tandem repeats with surrounding nucleotide sequences different from probe 2, probe 6 might form a more stable complex with Cdx4 protein. Similar to most homeodomain proteins, Cdx4 possibly forms a dimer to bind to target DNA –. Thus, two tandem sites may stabilize the Cdx4/ DNA interaction, resulting in supershift of Cdx4/DNA complex.
(A) A diagram of Hoxa9 locus, showing the corresponding locations between amplicons for qChIPs, reporter elements for luciferase analysis and oligo probes for EMSA.
(B) The wild-type and mutation sequences of probe 2 (left) and probe 6 (right).
The Cdx protein consensus binding sites were underlined.
Wild type sequence of Probe 2 contains one potential Cdx consensus binding site (CBSs) while probe 6 contains two tandem CBSs.
(C) Cdx4 protein was able to bind probe 2 wild type sequence but not the mutant sequence.
Asterisk (*) denotes a band that likely resulted from partially degraded Cdx4.
Their binding was inhibited by anti-Cdx4 antibody but not by a control IgG (left panel).
The same in vitro binding assay was conducted for probe 6 and its mutant (right panel).
To further evaluate the specificity of Cdx4 and probes 2 and 6 interaction, we introduced three nucleotide substitutions into the CBSs in each probe (Fig. 6B), and each of the mutated probes failed to bind Cdx4 (Fig. 6C, lanes 8–14, and 22–28). This series of experiments indicate that Cdx4 interacts in vitro with CBSs from element B and E (Fig. 5A), which confers Cdx4 direct upregulation to reporter expression.
Menin modulates Cdx4 access to the chromatin targets in Hoxa9 locus
To determine whether Cdx4 associates with the Hoxa9 locus in vivo, we performed qChIP assays. We found that menin and Cdx4 co-localized to the 5′ promoter region, and particularly enriched a region represented by amplicon 2 (Fig. 7A). Menin ablation abolished Cdx4-chromatin binding, while menin binding to the region was also markedly reduced due to menin excision from the cells (Fig.7A). Agarose electrophoresis of PCR products confirmed that the appropriate target was amplified (Fig. 7B). Unexpectedly, amplicon 4, which contains two CBSs was not as efficiently bound by Cdx4 as amplicon 2 (Fig. 7A). Thus, Cdx4 binding to the chromatin target depends on accessibility of the targets, and menin facilitates association of Cdx4 with the Hoxa9 locus.
(A) qChIP assay shows that Cdx4 and menin have a similar binding peak at amplicon 2.
Menin excision (AT1 Men1Δ/Δ) almost completely blocked Cdx4 binding.
(B) The amplified PCR product for amplicon 2 was separated on 1.5% agarose gel.
Hox genes play an essential role in controlling normal hematopoiesis as well as leukemogenesis , . The expression of Hox gene is tightly regulated during the process of hematopoietic proliferation and differentiation. For example, Hoxa9 and Hoxa7 are expressed in primitive progenitors and downregulated at the time of differentiation –. Histone modifications provide a dynamic and versatile code for the recruitment of factors that control gene transcription . We are the first to have demonstrated that Cdx4 directly bound the Hoxa9 locus and cooperated with menin to regulate Hoxa9 expression. Thus, our results reveal two distinct yet related mechanisms in regulating Hoxa9 expression in hematopoietic cells: Cdx4 binding to the CBSs and menin-dependent multiple H3K4 methylations at the Hoxa9 locus.
Menin-dependent H3K4 methylation is closely correlated with expression of the 5′ Hoxa cluster in MLL-AF9 transformed cells
Modern tumor biology has emphasized a quantitative model by which cells undergo gradual ‘progressive’ alterations, taking them from normal phenotype to abnormal (precancer) and to malignant phenotype (cancer) , . To this end, we applied a quantitative strategy to link the expression level of Hox genes with the enrichment of specific histone methylations in Hoxa9 regulatory region. This was performed in several cell lines with distinct levels of Hoxa9 expression in decreasing magnitude: AR1, AT1, and MEF26 cells. In correlation with an increase of Hoxa9 expression, an enriched peak of H3K4M3 was shifted from 3′ to 5′ of the Hoxa9 regulatory sequence from MEF26 to AR1 (Fig. 1). Trimethylated H3K4 has been reported to associate primarily with the promoter regions of actively expressed genes ,  and extended across the Hoxa9 locus in MLL-fusion transformed cells , . Moreover, our data suggest that the extension of trimethylated H3K4 towards the 5′ Hoxa9 regulatory region is closely correlated with increasing expression of Hoxa9. It is possible that H3K4M3 in the 5′ part of the locus is more efficient in recruiting cofactors, such as chd1 which activates gene transcription by chromatin remodeling to stimulate gene transcription , , , . Further studies are required to clarify this possibility. Despite loss of the SET domain in MLL fusion proteins, they can actively increase binding of the endogenous and intact MLL to the target genes 5 to 15 fold, leading to increased H3K4 trimethylation . Thus, one way that menin regulates expression of Hoxa9 and other Hoxa cluster genes may be to modulate the recruitment of endogenous MLL and MLL-AF9 to a particular locus to alter histone modifications.
Detection of the co-existence of active and repressive histone modifications at the Hoxa9 locus is consistent with the notion that certain genes controlling cell “stemness” or plurotency are coordinately regulated by active and repressive chromatin modifications , . Alternations of this co-existence may also influence the expression level of Hoxa9. For instance, in AT1 cells and MEF26 cells both active and repressive histone modification markers were enriched whereas, repressive histone modifications were significantly diminished in AR1 cells that express a high level of Hoxa9.
We further demonstrated that menin is a crucial regulator for both active and repressive histone 3 modifications in the Hoxa9 locus. Ablation of menin almost abolished both active and repressive histone modifications and reduced expression of 5′Hoxa cluster genes, Hoxa5 to a10 in AT1 cells. Menin associates with MLL and the Hoxa9 locus , , , . Moreover, our new findings suggest the possibility that menin may play a role in the communication between activator and repressor factors in Hox gene regulation. Further studies are needed to decipher this function.
Cdx4 binds the Hoxa9 locus and is required for Hoxa 9 expression in normal hematopoietic cells
Cdx4 is required for normal hematopoiesis, stem cell function, and Hoxa9 expression in both mice and Zebrafish , . Enforcing its crucial role, a point mutation of Zebrafish Cdx4 homolog abrogates normal hematopoiesis and Hoxa9 transcription, while ectopic expression of Hoxa9 can partially rescue the defect . However, it remains unclear whether Cdx4 directly binds any of Hox genes' loci to regulate their transcription. We have now shown that Cdx4 directly binds to the CBSs in the 5′ portion of the Hoxa9 locus, both in vitro and in vivo. Significantly, knockdown of Cdx4 led to marked reduction of multiple 5′ Hoxa cluster genes, including Hoxa9 in AT1 cells (Fig. 4B), whereas ectopic expression of Cdx4 upregulates expression of multiple endogenous Hoxa cluster genes including Hoxa9. These results indicate an essential role for Cdx4 in promoting expression of endogenous Hoxa genes. Further supporting these observations, Cdx4 upregulates expression of the reporter driven by the Hoxa9 elements containing CBSs, but not the reporter driven by elements without CBSs (Fig. 5). Together, these results indicate that Cdx4 can directly bind the Hoxa9 locus to promote transcription of Hoxa9, uncovering a direct link between Cdx4 and expression of Hox genes.
Cdx4 and menin co-regulate the expression of 5 Hoxa cluster genes
We have shown that Cdx4 and menin bound to the same region in the Hoxa9 locus, and they also additively activate the cis-element from the locus. Importantly, loss of menin prevents Cdx4 targeting the Hoxa9 gene. The precise nature of the interaction among menin, Cdx4 and chromatin DNA targets is not yet clear, but could involve the menin-MLL complex, which is known to mediate H3K4 methylation. However, we do not have evidence that Cdx4 directly interacts with menin in co-immunoprecipitation assay (A.S., data not shown). Histone modifications can directly affect chromatin structure and present a special surface for interaction with other proteins, resulting in either activation or repression of gene transcription . Although Cdx4 possesses intrinsic ability to bind the CBSs, Cdx4 itself is not sufficient to initiate Hoxa9 expression in the embryonic stem cells (data not shown). Menin-dependent histone modifications via MLL in the Hoxa9 locus may further recruit/stabilize Cdx4 in the chromatin and thus, establish an optimal level of Hoxa cluster gene expression at certain developmental stage of hematopoietic cells, such as AT1 cells. Obviously, menin is an essential oncogenic cofactor to regulate Hoxa9 expression in MLL-associated leukemogenesis  whereas Cdx4 is no longer required to maintain a high level of Hox gene expression in AR1 cells (Fig. 4B, 4D). Thus, Cdx4 may function as a temporal regulator in establishment and/or modulation of Hox expression. Indeed, Cdx proteins have long been recognized as playing an important role in the initiation of Hox gene expression  in a dosage-dependent mode .
Aberrant histone modifications are implicated in MLL-associated leukemogensis
MLL-AF9 promotes expansion of myeloid precursors, Hox gene dysregulation and the eventual development of myeloid leukemia , , . Transformation of MLL-AF9 fusion gene into mouse ES cells demonstrated that non-malignant expansion of myeloid precursors is the first stage of MLL-AF9-mediated leukemia followed by accumulation of malignant cells in bone marrow and other tissue . The long latency in vivo and monoclonal nature of the leukemias suggest that the secondary genetic alterations are required , , . Here, we provided additional evidence that dysregulation of 5′ Hoxa cluster is closely correlated with aberrant histone H3 modifications in Hoxa9 locus. In AR1 cells the expression of Hoxa5, a7 and a9 was increased 20 to 40 fold higher than in AT1 cells, and in particular, Hoxa10 expression was increased by 105 fold. In parallel with the dysregulation of 5′Hoxa cluster genes from a5 to a10, an enriched peak of trimethylated H3K4 was shifted from the 3′ to 5′ direction of the Hoxa9 regulatory sequence, and repressive histone modifications were suppressed in aggressive AR1 cells as compared to AT1 cells (Fig.1). Our data suggest that the “progressive” epigenetic alterations occurred from AT1 to AR1 possibly contribute to the secondary mutations that are required for the development of leukemia in the MLL-AF9 murine model system , . Taken together, our studies suggest a coordinated role for Cdx4 and the menin-MLL complex in modulating Hoxa9 expression during MLL-associated myeloid transformations, while Cdx4 is no longer required in full-blown leukemia cells such as AR1 cells. Comparison of sequences in UniGene with Cdx4 and menin proteins between mouse and human complete genome indicates that these proteins are highly conserved with amino acid identity of 82.39% in Cdx4 and 95.78% in menin. The DNA sequence covering regulatory elements B to F at Mouse Hoxa9 locus (Fig. 5A) possess 71% to 96% of DNA sequence identity to human Hoxa9 locus. The identities of the element B and element E which contain Cdx4 binding sites are 77% and 92%, respectively. Therefore, our studies have significant biological implications in further deciphering the detailed mechanisms whereby Hox genes are regulated to control hematopoiesis and leukemogenesis as well as devising novel therapeutic interventions by targeting the menin pathway in humans.
Materials and Methods
Cell lines and cell culture
The AT1 cell line was generated from the murine bone marrow cells of Men1l/l, Cre-ER mice by introduction with retrovirus expressing MLL-AF9 fusion protein, as we previously described . Similarly, the other MLL-AF9 cell line (AR1) was generated by transducing MLL-AF9 fusion protein into the bone marrow cells of Men1l/l mice. AT1 cells and AR1 cells were cultured in modified IMDM [Iscove's Modified Dulbecco's medium (StemCell Technologies Inc, Vancouver, BC) containing 200 mM non-essential amino acids, 200 mM L-glutamine, 10 U/ml penicillin and 10 µg/ml Streptomycin supplemented with 15% fetal bovine serum (FBS) (Harlan Bioproducts for Science, Indianapolis, IN) and IL3].
Retroviral constructs and infection
Retroviral MSCV-Cdx4-ires-GFP and MSCV-ires-GFP vectors were kindly provided by George Daley . The viruses were packaged to infect AT1 cells and AR1 cells as previously described . The retrovirally transduced cells by MSCV-Cdx4-GFP and MSCV-GFP virus were highly purified on the expression of GFP using MoFlo Cell Sorter (Dako Cytomation).
Quantitative Gene Expression Analysis
Total RNA was prepared with Trizol reagent, and cDNA was synthesized with random hexamers and Superscript II (Invitrogen). Taqman PCR reactions were performed with 1–2 µl of cDNA on a 7500 Fast Real-Time PCR System (Applied Biosystems) following the manufacturer's instructions. Primers and probes labeled at the 5′ end with FAM (MGB TaqMan Probe) were listed in the Supporting Information, Data S2. GAPDH (Mm99999915) and/or β-actin (Mm00607939) were used as internal controls for normalization. Data was analyzed using the comparative CT method following the manual described (Applied Biosystem). The average CT value from three independent assays was listed in the Supporting Information, Data S2.
DNA fragments (regulatory elements) from the promoter and 5′ UTR regions were obtained by PCR using mouse tail genomic DNA as the template and cloned into the pGL3-basic vector (Promega) at Kpn-BglII restriction sites. All constructs were validated by restriction digestions and sequencing analyses. Co-transfections of 500 ng luciferase reporters (pA, pB, pC, pD, pE, pF or pGL3 empty vector) and 1 µg of expression vectors were performed in 12 well plates using Lipofectamine 2000 (Invitrogen) with 50 ng of pTK-Renilla (Promega) as an internal control. The luciferase activities were measured on TR717 Microplate Luminometer (Tropix, PE Applied Biosystem) using the Dual-luciferase assay kit (Promega). The following expression vectors were utilized: pSG5-Cdx4 and pcDNA3-Menin, which were generated by inserting the mouse Cdx4 cDNA into pSG5  and the human menin cDNA into pcDNA3  respectively. The empty pcDNA3 was used as a vector control to normalize the total DNA for transfection.
Chromatin Immunoprecipitation (ChIP) analysis
Chromatin immunoprecipitation assays were performed using the Chromatin Immunoprecipitation Assay Kit (Upstate, Lake Pacid, NY) and specific antibodies (See Supporting Information S1). As control for ChIP assays and for antibody specificity, equal amounts (5 µg) of IgG and antibodies were used in parallel reactions. Using Taqman Real-time PCR (RT-PCR), the immunoprecipitated DNA samples were quantified relative to input DNA and calculated based on the formula: % totalsample = [2(ΔCTinput−ΔCTsample)×% totalinput] . To reduce background, ΔCT value is determined by subtracting the average CTIgG value from the average CTinput value or CTsample value. Primer and probe sequences were included in Supporting Information, Data S2.
Electrophoretic mobility shift assay (EMSA)
EMSA was carried out as previously described  with the following modifications. The coding region of Cdx4 was amplified from the pSG5-Cdx4 vector using primers (Cdx4F2, cacc gtc acc atg tat gga atc tgc and Cdx4R2, ttc aga aac tat gac ctg ctg tat c). The PCR product was cloned into the pCR4-TOPO vector (Invitrogen). The resultant plasmid, pCR4-Cdx4, was sequenced and restriction digested to determine its insert orientation. The TnT3 Quick coupled transcription-translation system (Promega) was used to produce Cdx4 protein in vitro, while the TnT7 Quick coupled transcription-translation system (Promega) was used to produce Cdx4-R as a control from the same plasmid (pCR4-Cdx4). The in vitro translated products were immunoblotted with anti-Cdx4 antibody. Six wild type sequences of double-stranded end-labeled oliogonucleotides covering the following regions were tested for Cdx4 and Cdx4-R protein binding by EMSA: probe1, probe 2 and probe 3 located within element B, probe 4 and probe 5 within element D and probe 6 within element E.
Whole-cell extracts were prepared followed by protein concentration determination. Equivalent amounts of protein from each sample were then subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) on a 10% gel, transferred to nitrocellulose, and processed for immunoblotting with the indicated antibodies .
Characterization of AT1 and AR1 cells. (A) AT1 and AR1 cells were seeded in 6-well plate in triplicate, 1×105 cells in each well. AR1 cells grow much faster than AT1 cells. (B) The expression profile of 5′Hoxa cluster genes in AT1 and AR1 cells was determined by quantitative real time RT-PCR analysis. Relative to AT1 cells, AR1 cells expressed significantly higher expression of Hoxa5, a7, a9 and a10, especially Hoxa10 with 105 fold of increase. (C) AT1 cells were predominately myeloblast morphology with a mixed population of a few differentiated myelomonocytes and myeloneutrophils, whereas, AR1 cells retained immature myelobalsts. Cells were stained with Giemsa and observed under light microscopy (Leisca, original magnification, ×40). Myeloblasts were morphologically defined by their large size, granular appearance and high nucleus-cytoplasm ratio , .
(12.81 MB TIF)
Identification of functional cis-elements at Hoxa9 regulatory region. (A) Diagram of Hoxa9 locus, showing locations of regulatory elements for luciferase assays. The regulatory elements were cloned into pGL3-basic vector to generate various reporter genes. (B) Luciferase assays were conducted in menin-expressing MEF26 cells. The luciferase activity was calculated relative to the empty vector control (pGL3-basic).
(4.74 MB TIF)
We are indebted to Shawn Burgess, Paul Liu, Zengfeng Wang and Martine Behra for critical reading of the manuscript. We thank all the other members of the Hua lab for stimulating discussions. We also thank Jacqueline Deschamps (Netherlands Institute for Developmental Biology, the Netherlands) for providing Cdx plasmids; George Daley for providing MSCV-Cdx4-ires-GFP and MSCV-ires-GFP vectors; and William DeMuth for FACS sorting GFP expressing cells. We acknowledge Douglas Lin, AFCRI's Diehl lab, for his expertise in shRNA technique.
Conceived and designed the experiments: XH JY. Performed the experiments: JY YC AD AS YY. Analyzed the data: XH JY YC. Contributed reagents/materials/analysis tools: HW. Wrote the paper: XH JY.
- 1. Magli MC, Largman C, Lawrence HJ (1997) Effects of HOX homeobox genes in blood cell differentiation. J Cell Physiol 173: 168–177.MC MagliC. LargmanHJ Lawrence1997Effects of HOX homeobox genes in blood cell differentiation.J Cell Physiol173168177
- 2. Owens BM, Hawley RG (2002) HOX and non-HOX homeobox genes in leukemic hematopoiesis. Stem Cells 20: 364–379.BM OwensRG Hawley2002HOX and non-HOX homeobox genes in leukemic hematopoiesis.Stem Cells20364379
- 3. Ayton PM, Cleary ML (2001) Molecular mechanisms of leukemogenesis mediated by MLL fusion proteins. Oncogene 20: 5695–5707.PM AytonML Cleary2001Molecular mechanisms of leukemogenesis mediated by MLL fusion proteins.Oncogene2056955707
- 4. Dobson CL, Warren AJ, Pannell R, Forster A, Lavenir I, et al. (1999) The mll-AF9 gene fusion in mice controls myeloproliferation and specifies acute myeloid leukaemogenesis. Embo J 18: 3564–3574.CL DobsonAJ WarrenR. PannellA. ForsterI. Lavenir1999The mll-AF9 gene fusion in mice controls myeloproliferation and specifies acute myeloid leukaemogenesis.Embo J1835643574
- 5. Look AT (1997) Oncogenic transcription factors in the human acute leukemias. Science 278: 1059–1064.AT Look1997Oncogenic transcription factors in the human acute leukemias.Science27810591064
- 6. Sorensen PH, Chen CS, Smith FO, Arthur DC, Domer PH, et al. (1994) Molecular rearrangements of the MLL gene are present in most cases of infant acute myeloid leukemia and are strongly correlated with monocytic or myelomonocytic phenotypes. J Clin Invest 93: 429–437.PH SorensenCS ChenFO SmithDC ArthurPH Domer1994Molecular rearrangements of the MLL gene are present in most cases of infant acute myeloid leukemia and are strongly correlated with monocytic or myelomonocytic phenotypes.J Clin Invest93429437
- 7. Chandrasekharappa SC, Guru SC, Manickam P, Olufemi SE, Collins FS, et al. (1997) Positional cloning of the gene for multiple endocrine neoplasia-type 1. Science 276: 404–407.SC ChandrasekharappaSC GuruP. ManickamSE OlufemiFS Collins1997Positional cloning of the gene for multiple endocrine neoplasia-type 1.Science276404407
- 8. Lemmens I, Van de Ven WJ, Kas K, Zhang CX, Giraud S, et al. (1997) Identification of the multiple endocrine neoplasia type 1 (MEN1) gene. The European Consortium on MEN1. Hum Mol Genet 6: 1177–1183.I. LemmensWJ Van de VenK. KasCX ZhangS. Giraud1997Identification of the multiple endocrine neoplasia type 1 (MEN1) gene. The European Consortium on MEN1.Hum Mol Genet611771183
- 9. Hughes CM, Rozenblatt-Rosen O, Milne TA, Copeland TD, Levine SS, et al. (2004) Menin associates with a trithorax family histone methyltransferase complex and with the hoxc8 locus. Mol Cell 13: 587–597.CM HughesO. Rozenblatt-RosenTA MilneTD CopelandSS Levine2004Menin associates with a trithorax family histone methyltransferase complex and with the hoxc8 locus.Mol Cell13587597
- 10. Lemmens IH, Forsberg L, Pannett AA, Meyen E, Piehl F, et al. (2001) Menin interacts directly with the homeobox-containing protein Pem. Biochem Biophys Res Commun 286: 426–431.IH LemmensL. ForsbergAA PannettE. MeyenF. Piehl2001Menin interacts directly with the homeobox-containing protein Pem.Biochem Biophys Res Commun286426431
- 11. Lin SY, Elledge SJ (2003) Multiple tumor suppressor pathways negatively regulate telomerase. Cell 113: 881–889.SY LinSJ Elledge2003Multiple tumor suppressor pathways negatively regulate telomerase.Cell113881889
- 12. Yokoyama A, Wang Z, Wysocka J, Sanyal M, Aufiero DJ, et al. (2004) Leukemia proto-oncoprotein MLL forms a SET1-like histone methyltransferase complex with menin to regulate Hox gene expression. Mol Cell Biol 24: 5639–5649.A. YokoyamaZ. WangJ. WysockaM. SanyalDJ Aufiero2004Leukemia proto-oncoprotein MLL forms a SET1-like histone methyltransferase complex with menin to regulate Hox gene expression.Mol Cell Biol2456395649
- 13. Chen YX, Yan J, Keeshan K, Tubbs AT, Wang H, et al. (2006) The tumor suppressor menin regulates hematopoiesis and myeloid transformation by influencing Hox gene expression. Proc Natl Acad Sci U S A 103: 1018–1023.YX ChenJ. YanK. KeeshanAT TubbsH. Wang2006The tumor suppressor menin regulates hematopoiesis and myeloid transformation by influencing Hox gene expression.Proc Natl Acad Sci U S A10310181023
- 14. Yokoyama A, Somervaille TC, Smith KS, Rozenblatt-Rosen O, Meyerson M, et al. (2005) The Menin Tumor Suppressor Protein Is an Essential Oncogenic Cofactor for MLL-Associated Leukemogenesis. Cell 123: 207–218.A. YokoyamaTC SomervailleKS SmithO. Rozenblatt-RosenM. Meyerson2005The Menin Tumor Suppressor Protein Is an Essential Oncogenic Cofactor for MLL-Associated Leukemogenesis.Cell123207218
- 15. Pollard SL, Holland PW (2000) Evidence for 14 homeobox gene clusters in human genome ancestry. Curr Biol 10: 1059–1062.SL PollardPW Holland2000Evidence for 14 homeobox gene clusters in human genome ancestry.Curr Biol1010591062
- 16. Davidson AJ, Ernst P, Wang Y, Dekens MP, Kingsley PD, et al. (2003) cdx4 mutants fail to specify blood progenitors and can be rescued by multiple hox genes. Nature 425: 300–306.AJ DavidsonP. ErnstY. WangMP DekensPD Kingsley2003cdx4 mutants fail to specify blood progenitors and can be rescued by multiple hox genes.Nature425300306
- 17. Lawrence HJ, Christensen J, Fong S, Hu YL, Weissman I, et al. (2005) Loss of expression of the Hoxa-9 homeobox gene impairs the proliferation and repopulating ability of hematopoietic stem cells. Blood 106: 3988–3994.HJ LawrenceJ. ChristensenS. FongYL HuI. Weissman2005Loss of expression of the Hoxa-9 homeobox gene impairs the proliferation and repopulating ability of hematopoietic stem cells.Blood10639883994
- 18. Ayton PM, Cleary ML (2003) Transformation of myeloid progenitors by MLL oncoproteins is dependent on Hoxa7 and Hoxa9. Genes Dev 17: 2298–2307.PM AytonML Cleary2003Transformation of myeloid progenitors by MLL oncoproteins is dependent on Hoxa7 and Hoxa9.Genes Dev1722982307
- 19. Calvo KR, Sykes DB, Pasillas M, Kamps MP (2000) Hoxa9 immortalizes a granulocyte-macrophage colony-stimulating factor-dependent promyelocyte capable of biphenotypic differentiation to neutrophils or macrophages, independent of enforced meis expression. Mol Cell Biol 20: 3274–3285.KR CalvoDB SykesM. PasillasMP Kamps2000Hoxa9 immortalizes a granulocyte-macrophage colony-stimulating factor-dependent promyelocyte capable of biphenotypic differentiation to neutrophils or macrophages, independent of enforced meis expression.Mol Cell Biol2032743285
- 20. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, et al. (1999) Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 286: 531–537.TR GolubDK SlonimP. TamayoC. HuardM. Gaasenbeek1999Molecular classification of cancer: class discovery and class prediction by gene expression monitoring.Science286531537
- 21. Nakamura T, Largaespada DA, Shaughnessy JD Jr., Jenkins NA, Copeland NG (1996) Cooperative activation of Hoxa and Pbx1-related genes in murine myeloid leukaemias. Nat Genet 12: 149–153.T. NakamuraDA LargaespadaJD Shaughnessy Jr.NA JenkinsNG Copeland1996Cooperative activation of Hoxa and Pbx1-related genes in murine myeloid leukaemias.Nat Genet12149153
- 22. Johnson JJ, Chen W, Hudson W, Yao Q, Taylor M, et al. (2003) Prenatal and postnatal myeloid cells demonstrate stepwise progression in the pathogenesis of MLL fusion gene leukemia. Blood 101: 3229–3235.JJ JohnsonW. ChenW. HudsonQ. YaoM. Taylor2003Prenatal and postnatal myeloid cells demonstrate stepwise progression in the pathogenesis of MLL fusion gene leukemia.Blood10132293235
- 23. Kumar AR, Hudson WA, Chen W, Nishiuchi R, Yao Q, et al. (2004) Hoxa9 influences the phenotype but not the incidence of Mll-AF9 fusion gene leukemia. Blood 103: 1823–1828.AR KumarWA HudsonW. ChenR. NishiuchiQ. Yao2004Hoxa9 influences the phenotype but not the incidence of Mll-AF9 fusion gene leukemia.Blood10318231828
- 24. Dou Y, Milne TA, Tackett AJ, Smith ER, Fukuda A, et al. (2005) Physical association and coordinate function of the H3 K4 methyltransferase MLL1 and the H4 K16 acetyltransferase MOF. Cell 121: 873–885.Y. DouTA MilneAJ TackettER SmithA. Fukuda2005Physical association and coordinate function of the H3 K4 methyltransferase MLL1 and the H4 K16 acetyltransferase MOF.Cell121873885
- 25. Schnepp RW, Hou Z, Wang H, Petersen C, Silva A, et al. (2004) Functional interaction between tumor suppressor menin and activator of S-phase kinase. Cancer Res 64: 6791–6796.RW SchneppZ. HouH. WangC. PetersenA. Silva2004Functional interaction between tumor suppressor menin and activator of S-phase kinase.Cancer Res6467916796
- 26. Deschamps J, van Nes J (2005) Developmental regulation of the Hox genes during axial morphogenesis in the mouse. Development 132: 2931–2942.J. DeschampsJ. van Nes2005Developmental regulation of the Hox genes during axial morphogenesis in the mouse.Development13229312942
- 27. Cao R, Zhang Y (2004) SUZ12 is required for both the histone methyltransferase activity and the silencing function of the EED-EZH2 complex. Mol Cell 15: 57–67.R. CaoY. Zhang2004SUZ12 is required for both the histone methyltransferase activity and the silencing function of the EED-EZH2 complex.Mol Cell155767
- 28. Cao R, Wang L, Wang H, Xia L, Erdjument-Bromage H, et al. (2002) Role of histone H3 lysine 27 methylation in Polycomb-group silencing. Science 298: 1039–1043.R. CaoL. WangH. WangL. XiaH. Erdjument-Bromage2002Role of histone H3 lysine 27 methylation in Polycomb-group silencing.Science29810391043
- 29. Milne TA, Briggs SD, Brock HW, Martin ME, Gibbs D, et al. (2002) MLL targets SET domain methyltransferase activity to Hox gene promoters. Mol Cell 10: 1107–1117.TA MilneSD BriggsHW BrockME MartinD. Gibbs2002MLL targets SET domain methyltransferase activity to Hox gene promoters.Mol Cell1011071117
- 30. Dehe PM, Pamblanco M, Luciano P, Lebrun R, Moinier D, et al. (2005) Histone H3 lysine 4 mono-methylation does not require ubiquitination of histone H2B. J Mol Biol 353: 477–484.PM DeheM. PamblancoP. LucianoR. LebrunD. Moinier2005Histone H3 lysine 4 mono-methylation does not require ubiquitination of histone H2B.J Mol Biol353477484
- 31. Morillon A, Karabetsou N, Nair A, Mellor J (2005) Dynamic lysine methylation on histone H3 defines the regulatory phase of gene transcription. Mol Cell 18: 723–734.A. MorillonN. KarabetsouA. NairJ. Mellor2005Dynamic lysine methylation on histone H3 defines the regulatory phase of gene transcription.Mol Cell18723734
- 32. van Dijk K, Marley KE, Jeong BR, Xu J, Hesson J, et al. (2005) Monomethyl histone H3 lysine 4 as an epigenetic mark for silenced euchromatin in Chlamydomonas. Plant Cell 17: 2439–2453.K. van DijkKE MarleyBR JeongJ. XuJ. Hesson2005Monomethyl histone H3 lysine 4 as an epigenetic mark for silenced euchromatin in Chlamydomonas.Plant Cell1724392453
- 33. Lachner M, Jenuwein T (2002) The many faces of histone lysine methylation. Curr Opin Cell Biol 14: 286–298.M. LachnerT. Jenuwein2002The many faces of histone lysine methylation.Curr Opin Cell Biol14286298
- 34. Nakamura T, Mori T, Tada S, Krajewski W, Rozovskaia T, et al. (2002) ALL-1 is a histone methyltransferase that assembles a supercomplex of proteins involved in transcriptional regulation. Mol Cell 10: 1119–1128.T. NakamuraT. MoriS. TadaW. KrajewskiT. Rozovskaia2002ALL-1 is a histone methyltransferase that assembles a supercomplex of proteins involved in transcriptional regulation.Mol Cell1011191128
- 35. Wang Y, Yates F, Naveiras O, Ernst P, Daley GQ (2005) Embryonic stem cell-derived hematopoietic stem cells. Proc Natl Acad Sci U S A 102: 19081–19086.Y. WangF. YatesO. NaveirasP. ErnstGQ Daley2005Embryonic stem cell-derived hematopoietic stem cells.Proc Natl Acad Sci U S A1021908119086
- 36. Tabaries S, Lapointe J, Besch T, Carter M, Woollard J, et al. (2005) Cdx protein interaction with Hoxa5 regulatory sequences contributes to Hoxa5 regional expression along the axial skeleton. Mol Cell Biol 25: 1389–1401.S. TabariesJ. LapointeT. BeschM. CarterJ. Woollard2005Cdx protein interaction with Hoxa5 regulatory sequences contributes to Hoxa5 regional expression along the axial skeleton.Mol Cell Biol2513891401
- 37. Gregory PA, Gardner-Stephen DA, Rogers A, Michael MZ, Mackenzie PI (2006) The caudal-related homeodomain protein Cdx2 and hepatocyte nuclear factor 1alpha cooperatively regulate the UDP-glucuronosyltransferase 2B7 gene promoter. Pharmacogenet Genomics 16: 527–536.PA GregoryDA Gardner-StephenA. RogersMZ MichaelPI Mackenzie2006The caudal-related homeodomain protein Cdx2 and hepatocyte nuclear factor 1alpha cooperatively regulate the UDP-glucuronosyltransferase 2B7 gene promoter.Pharmacogenet Genomics16527536
- 38. Mendel DB, Khavari PA, Conley PB, Graves MK, Hansen LP, et al. (1991) Characterization of a cofactor that regulates dimerization of a mammalian homeodomain protein. Science 254: 1762–1767.DB MendelPA KhavariPB ConleyMK GravesLP Hansen1991Characterization of a cofactor that regulates dimerization of a mammalian homeodomain protein.Science25417621767
- 39. Wilson DS, Guenther B, Desplan C, Kuriyan J (1995) High resolution crystal structure of a paired (Pax) class cooperative homeodomain dimer on DNA. Cell 82: 709–719.DS WilsonB. GuentherC. DesplanJ. Kuriyan1995High resolution crystal structure of a paired (Pax) class cooperative homeodomain dimer on DNA.Cell82709719
- 40. Moore MA (2005) Converging pathways in leukemogenesis and stem cell self-renewal. Exp Hematol 33: 719–737.MA Moore2005Converging pathways in leukemogenesis and stem cell self-renewal.Exp Hematol33719737
- 41. Kawagoe H, Humphries RK, Blair A, Sutherland HJ, Hogge DE (1999) Expression of HOX genes, HOX cofactors, and MLL in phenotypically and functionally defined subpopulations of leukemic and normal human hematopoietic cells. Leukemia 13: 687–698.H. KawagoeRK HumphriesA. BlairHJ SutherlandDE Hogge1999Expression of HOX genes, HOX cofactors, and MLL in phenotypically and functionally defined subpopulations of leukemic and normal human hematopoietic cells.Leukemia13687698
- 42. Lawrence HJ, Rozenfeld S, Cruz C, Matsukuma K, Kwong A, et al. (1999) Frequent co-expression of the HOXA9 and MEIS1 homeobox genes in human myeloid leukemias. Leukemia 13: 1993–1999.HJ LawrenceS. RozenfeldC. CruzK. MatsukumaA. Kwong1999Frequent co-expression of the HOXA9 and MEIS1 homeobox genes in human myeloid leukemias.Leukemia1319931999
- 43. Berger SL (2002) Histone modifications in transcriptional regulation. Curr Opin Genet Dev 12: 142–148.SL Berger2002Histone modifications in transcriptional regulation.Curr Opin Genet Dev12142148
- 44. Araten DJ, Golde DW, Zhang RH, Thaler HT, Gargiulo L, et al. (2005) A quantitative measurement of the human somatic mutation rate. Cancer Res 65: 8111–8117.DJ AratenDW GoldeRH ZhangHT ThalerL. Gargiulo2005A quantitative measurement of the human somatic mutation rate.Cancer Res6581118117
- 45. Luzzatto L (2001) The mechanisms of neoplastic transformation. Eur J Cancer 37: Suppl 8S114–117.L. Luzzatto2001The mechanisms of neoplastic transformation.Eur J Cancer37Suppl 8S114117
- 46. Bernstein BE, Kamal M, Lindblad-Toh K, Bekiranov S, Bailey DK, et al. (2005) Genomic maps and comparative analysis of histone modifications in human and mouse. Cell 120: 169–181.BE BernsteinM. KamalK. Lindblad-TohS. BekiranovDK Bailey2005Genomic maps and comparative analysis of histone modifications in human and mouse.Cell120169181
- 47. Schneider R, Bannister AJ, Myers FA, Thorne AW, Crane-Robinson C, et al. (2004) Histone H3 lysine 4 methylation patterns in higher eukaryotic genes. Nat Cell Biol 6: 73–77.R. SchneiderAJ BannisterFA MyersAW ThorneC. Crane-Robinson2004Histone H3 lysine 4 methylation patterns in higher eukaryotic genes.Nat Cell Biol67377
- 48. Milne TA, Martin ME, Brock HW, Slany RK, Hess JL (2005) Leukemogenic MLL fusion proteins bind across a broad region of the Hox a9 locus, promoting transcription and multiple histone modifications. Cancer Res 65: 11367–11374.TA MilneME MartinHW BrockRK SlanyJL Hess2005Leukemogenic MLL fusion proteins bind across a broad region of the Hox a9 locus, promoting transcription and multiple histone modifications.Cancer Res651136711374
- 49. Krogan NJ, Kim M, Tong A, Golshani A, Cagney G, et al. (2003) Methylation of histone H3 by Set2 in Saccharomyces cerevisiae is linked to transcriptional elongation by RNA polymerase II. Mol Cell Biol 23: 4207–4218.NJ KroganM. KimA. TongA. GolshaniG. Cagney2003Methylation of histone H3 by Set2 in Saccharomyces cerevisiae is linked to transcriptional elongation by RNA polymerase II.Mol Cell Biol2342074218
- 50. Pray-Grant MG, Daniel JA, Schieltz D, Yates JR 3rd, Grant PA (2005) Chd1 chromodomain links histone H3 methylation with SAGA- and SLIK-dependent acetylation. Nature 433: 434–438.MG Pray-GrantJA DanielD. SchieltzJR Yates 3rdPA Grant2005Chd1 chromodomain links histone H3 methylation with SAGA- and SLIK-dependent acetylation.Nature433434438
- 51. Lee TI, Jenner RG, Boyer LA, Guenther MG, Levine SS, et al. (2006) Control of developmental regulators by Polycomb in human embryonic stem cells. Cell 125: 301–313.TI LeeRG JennerLA BoyerMG GuentherSS Levine2006Control of developmental regulators by Polycomb in human embryonic stem cells.Cell125301313
- 52. Yokoyama A, Kitabayashi I, Ayton PM, Cleary ML, Ohki M (2002) Leukemia proto-oncoprotein MLL is proteolytically processed into 2 fragments with opposite transcriptional properties. Blood 100: 3710–3718.A. YokoyamaI. KitabayashiPM AytonML ClearyM. Ohki2002Leukemia proto-oncoprotein MLL is proteolytically processed into 2 fragments with opposite transcriptional properties.Blood10037103718
- 53. Gaunt SJ, Drage D, Trubshaw RC (2005) cdx4/lacZ and cdx2/lacZ protein gradients formed by decay during gastrulation in the mouse. Int J Dev Biol 49: 901–908.SJ GauntD. DrageRC Trubshaw2005cdx4/lacZ and cdx2/lacZ protein gradients formed by decay during gastrulation in the mouse.Int J Dev Biol49901908
- 54. Martin ME, Milne TA, Bloyer S, Galoian K, Shen W, et al. (2003) Dimerization of MLL fusion proteins immortalizes hematopoietic cells. Cancer Cell 4: 197–207.ME MartinTA MilneS. BloyerK. GaloianW. Shen2003Dimerization of MLL fusion proteins immortalizes hematopoietic cells.Cancer Cell4197207
- 55. Lavau C, Du C, Thirman M, Zeleznik-Le N (2000) Chromatin-related properties of CBP fused to MLL generate a myelodysplastic-like syndrome that evolves into myeloid leukemia. Embo J 19: 4655–4664.C. LavauC. DuM. ThirmanN. Zeleznik-Le2000Chromatin-related properties of CBP fused to MLL generate a myelodysplastic-like syndrome that evolves into myeloid leukemia.Embo J1946554664
- 56. Lavau C, Luo RT, Du C, Thirman MJ (2000) Retrovirus-mediated gene transfer of MLL-ELL transforms primary myeloid progenitors and causes acute myeloid leukemias in mice. Proc Natl Acad Sci U S A 97: 10984–10989.C. LavauRT LuoC. DuMJ Thirman2000Retrovirus-mediated gene transfer of MLL-ELL transforms primary myeloid progenitors and causes acute myeloid leukemias in mice.Proc Natl Acad Sci U S A971098410989
- 57. Jin S, Mao H, Schnepp RW, Sykes SM, Silva AC, et al. (2003) Menin associates with FANCD2, a protein involved in repair of DNA damage. Cancer Res 63: 4204–4210.S. JinH. MaoRW SchneppSM SykesAC Silva2003Menin associates with FANCD2, a protein involved in repair of DNA damage.Cancer Res6342044210
- 58. Frank SR, Schroeder M, Fernandez P, Taubert S, Amati B (2001) Binding of c-Myc to chromatin mediates mitogen-induced acetylation of histone H4 and gene activation. Genes Dev 15: 2069–2082.SR FrankM. SchroederP. FernandezS. TaubertB. Amati2001Binding of c-Myc to chromatin mediates mitogen-induced acetylation of histone H4 and gene activation.Genes Dev1520692082
- 59. Schnepp RW, Mao H, Sykes SM, Zong WX, Silva A, et al. (2004) Menin induces apoptosis in murine embryonic fibroblasts. J Biol Chem 279: 10685–10691.RW SchneppH. MaoSM SykesWX ZongA. Silva2004Menin induces apoptosis in murine embryonic fibroblasts.J Biol Chem2791068510691